Pregabalin: Independent Evidence on Pain, Anxiety, Side Effects & Withdrawal

Table of contents

• Evidence summary
• What pregabalin is
• How it works
• What it is prescribed for
• What works and what does not
• Benefits by claim
• Risks and all side effects
• All interactions
• Who should avoid pregabalin
• Dosage and how to take it
• Follow the money: who makes it and who funded the evidence
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

What pregabalin is

Pregabalin is an antiepileptic (antiseizure) medicine. It belongs to the gabapentinoid class together with gabapentin (Neurontin). Its structure resembles the brain chemical GABA, but it does not act like GABA (EMA EPAR). The original brand is Lyrica. UK brand names also include Alzain, Axalid and Misabri, and it comes as capsules, tablets, slow-release tablets and a liquid (NHS). The MHRA also lists Lecaent among UK brands (MHRA 2022). The extended-release US version is Lyrica CR (FDA 2019).

Approval. The European Commission authorised Lyrica across the EU for Pfizer Limited on 6 July 2004 (EMA). The FDA says pregabalin "was first approved in 2004" in the US (FDA 2019).

Who developed it. The molecule was created at Northwestern University (Illinois, USA) in the laboratory of chemist Richard Silverman and licensed to what became part of Pfizer (The Daily Northwestern, 2015). The original UK patent holder, Warner-Lambert Company LLC, is "part of the Pfizer group of companies" (UK Supreme Court, 2018). The US label for the original Lyrica was issued by Parke-Davis, a division of Pfizer Inc. The current US Lyrica label is held by Viatris Specialty LLC (DailyMed label). The EU marketing authorisation holder is now Upjohn EESV, based in the Netherlands (EU product information).

Generic status. EU regulatory exclusivity ended in July 2014. In the US, multi-source generic competition began in July 2019 (Pfizer 2019 Financial Report). Cheap generics are now widely available.

Legal status.

• UK: prescription only. Since 1 April 2019 it has been a Class C controlled substance and a Schedule 3 drug, and it is illegal to have it without a prescription or to sell or supply it to others (MHRA 2019). Pharmacists ask for proof of identity when you collect it, and a prescription is valid for only 28 days (NHS).
• US: prescription only. It is a Schedule V controlled substance, the lowest schedule, which the FDA describes as having "a lower potential for abuse" but possibly leading "to some physical or psychological dependence" (FDA 2019).

How it works

Pregabalin binds tightly to the alpha-2-delta subunit, a helper part of voltage-gated calcium channels in the brain and spinal cord. The FDA label says its mechanism "has not been fully elucidated". In animal models of nerve damage it reduces calcium-dependent release of pain-signalling neurotransmitters in the spinal cord, and it may also act through descending noradrenergic and serotonergic pathways from the brainstem (FDA label, section 12.1). The EMA summarises it as affecting "the way that calcium enters nerve cells", which reduces the release of neurotransmitters involved in pain, epilepsy and anxiety (EMA).

How the body handles it explains several practical points:

• It is barely broken down by the liver. Less than 2% of a dose is recovered as metabolites, and it does not bind to plasma proteins, so it has few chemical drug interactions (FDA label, section 7).
• It leaves the body almost entirely through the kidneys. Its clearance is directly proportional to kidney function, so people with reduced kidney function, including many older adults, need lower doses (MHRA 2021).
• Its most dangerous interactions are additive. Its sedating effect stacks with opioids, benzodiazepines and alcohol. The FDA said its 2019 review "provides some evidence contrary to the widely held belief that gabapentinoids lack drug interactions and have wide therapeutic indices" (FDA 2019).

What it is prescribed for

The licensed uses differ between the UK/EU and the US, which is one of the more instructive things about this medicine.

Licence sources: MHRA 2021 (UK indications), FDA label, section 1, EMA refusal Q&A.

What works and what does not

Benefits by claim

Nerve pain: real, but for a minority

The 2019 Cochrane review is the most complete independent summary. It covers 45 randomised, double-blind studies with 11,906 participants, lasting 2 to 16 weeks (Derry et al., Cochrane 2019). Its main results, as reported:

• Post-herpetic neuralgia, 300 mg/day: at least 30% pain reduction in 50% against 25% on placebo (RR 2.1, NNT 3.9). At least 50% reduction in 32% against 13% (RR 2.5, NNT 5.3).
• Post-herpetic neuralgia, 600 mg/day: at least 30% reduction in 62% against 24% (NNT 2.7). At least 50% reduction in 41% against 15% (NNT 3.9).
• Painful diabetic neuropathy, 300 mg/day: at least 30% reduction in 47% against 42% (NNT 22). At least 50% reduction in 31% against 24% (NNT 22). "Much or very much improved" on the patient global impression scale in 51% against 30% (NNT 4.9).
• Painful diabetic neuropathy, 600 mg/day: at least 50% reduction in 41% against 28% (NNT 7.8; low-quality evidence).
• Serious adverse events: no more common than with placebo (high-quality evidence).

The authors' own conclusion is balanced: "Some people will derive substantial benefit with pregabalin; more will have moderate benefit, but many will have no benefit or will discontinue treatment." A number needed to treat of 22 means that roughly 22 people have to take the drug for one extra person to get half their pain relieved, compared with placebo. The placebo response is large in pain trials: about a quarter of people on dummy tablets in diabetic neuropathy studies reported at least 50% relief. A person who feels better on pregabalin may partly be experiencing that.

The review also flagged a publication problem. Completed but unreported studies held data on 2,098 participants, 1,829 of them in painful diabetic neuropathy, and the authors wrote that this "may have the potential to substantially alter the results reported here" (full review, PMC). The study tables record Pfizer sponsorship or funding for most of the included trials.

The independent test outside the licensed conditions was negative. PRECISE, a trial funded by the Australian government, randomised 209 people with sciatica to pregabalin (up to 600 mg/day) or placebo. At 8 weeks the average leg pain was 3.7 on pregabalin and 3.1 on placebo (adjusted difference 0.5, not significant), with no benefit at one year and more adverse events (Mathieson et al., NEJM 2017).

Generalised anxiety disorder: a moderate effect with a dependence trade-off

Slee and colleagues pooled 89 trials with 25,441 patients. Pregabalin reduced the Hamilton Anxiety Scale score by 2.79 points more than placebo (95% credible interval −3.69 to −1.91). This was in the same range as duloxetine (−3.13), venlafaxine (−2.69) and escitalopram (−2.45), and all four had "relatively good acceptability" (Slee et al., Lancet 2019). The HAM-A scale runs from 0 to 56, so an average difference of under 3 points is modest. A separate meta-analysis of eight trials (2,299 patients) found a small-to-moderate effect size (Hedges' g 0.37) (Generoso et al., 2017). The EMA's summary of eight studies with over 3,000 patients reported that 52% on pregabalin improved by at least half, against 38% on placebo (EMA). The placebo response is therefore large, and the drug adds roughly 14 percentage points on top of it.

NICE does not rank pregabalin first for GAD despite this efficacy. It suggests sertraline first, then another SSRI or an SNRI, and only then pregabalin if those cannot be tolerated. The recommendation carries the Class C and dependence warning, and NICE notes there is no evidence that drug treatment is better than high-intensity psychological therapy (NICE CG113). The FDA never approved pregabalin for GAD (FDA label). Our reviews of sertraline, escitalopram, duloxetine and venlafaxine cover the preferred alternatives.

Epilepsy: an effective add-on, evidence entirely from the manufacturer

In drug-resistant focal epilepsy, the 2022 Cochrane review included 11 trials with 3,949 participants. People given add-on pregabalin were more likely to have their seizures halved (RR 1.95, 95% CI 1.40 to 2.72; low certainty) and to become seizure-free (RR 3.94; moderate certainty). The odds of response doubled from 300 to 600 mg/day, but withdrawals because of adverse effects were also higher (RR 2.60). In head-to-head trials pregabalin was better than lamotrigine for seizure halving, similar to levetiracetam and gabapentin, and worse than levetiracetam for seizure freedom. The review rated every trial at "high risk of funding bias as they were all sponsored by Pfizer" (Panebianco et al., Cochrane 2022). The EMA summary reports that about 45% on 600 mg and 35% on 300 mg halved their seizures, against about 10% on placebo (EMA).

Risks and all side effects

Pregabalin has no FDA boxed warning. Its label does carry warnings on angioedema, hypersensitivity, suicidal thoughts, problems when it is stopped abruptly, respiratory depression, dizziness and sleepiness, peripheral oedema, weight gain, tumorigenic potential (a finding in mouse studies), eye effects, creatine kinase rises, low platelets and PR-interval prolongation (FDA label, sections 5–6). The FDA label reports these rates from controlled trials in adults:

• Dizziness: 30% on pregabalin against 8% on placebo.
• Somnolence (sleepiness): 23% against 8%.
• Stopped early because of adverse reactions: 14% against 7%.
• Dizziness and somnolence often lasted: in patients who reported them, dizziness was still present at the last dose in 30% and somnolence in 42%.

Why the UK made pregabalin a controlled drug

In 2016, after "concerns about misuse, illegal diversion, and dependence", the Advisory Council on the Misuse of Drugs recommended controlling both gabapentinoids. The Home Office accepted this, and the change took effect on 1 April 2019. The MHRA asked prescribers to "evaluate patients carefully for a history of drug abuse" and to watch for "drug-seeking behaviour, dose escalation, and development of tolerance" (MHRA 2019). Public Health England and NHS England had already issued advice to prescribers on misuse risk in December 2014 (PHE / NHS England 2014). An academic systematic review of 59 studies found gabapentinoid misuse in 1.6% of the general population, but between 3% and 68% among people who misuse opioids. It listed a history of substance misuse, especially of opioids, as the main risk factor (Evoy et al., Drugs 2017).

The opioid combination

This is the most dangerous practical risk. The MHRA states that "use of high doses of pregabalin (over 300mg a day) alongside opioid medicines" is "particularly associated with an increased risk of opioid-related death" (MHRA 2021). It cites a Canadian nested case-control study (Gomes et al., Ann Intern Med 2018). The same Ontario research group had earlier found that adding gabapentin to opioids raised the adjusted odds of opioid-related death by 49% (aOR 1.49) (Gomes et al., PLoS Med 2017). The 2023 Beers Criteria advise older adults to avoid combining opioids with gabapentin or pregabalin, because of "increased risk of severe sedation-related adverse events, including respiratory depression and death" (AGS Beers 2023). The exceptions are when someone is moving from an opioid to a gabapentinoid, or when a gabapentinoid is used to reduce the opioid dose.

All interactions

Pregabalin has almost no metabolic interactions, because the liver barely processes it. Its dangerous interactions come from other drugs that also depress the brain and breathing (FDA label, section 7).

Who should avoid pregabalin

• Anyone allergic to pregabalin or its ingredients. This is the only formal contraindication (FDA label).
• People with a history of drug or alcohol misuse, especially opioid misuse. They need careful evaluation because they are at higher risk of pregabalin misuse and dependence (EU SmPC, Evoy 2017).
• People taking opioids or other sedatives, or with lung disease such as COPD, neurological disease or kidney impairment. They have a higher risk of respiratory depression and may need lower doses (MHRA 2021).
• Older adults. The NHS says it "might not be suitable for people older than 65" (NHS). The 2023 Beers Criteria say to reduce the dose when creatinine clearance is below 60 mL/min, to avoid combining it with opioids, and to avoid using 3 or more CNS-active drugs together because of falls and fractures (AGS Beers 2023).
• Kidney impairment. The dose must be individualised to creatinine clearance, and haemodialysis removes about 50% in 4 hours (EU SmPC 4.2).
• Liver disease and heart disease. The NHS asks people to tell their doctor about either before starting (NHS). The FDA advises caution in severe (NYHA class III–IV) heart failure (FDA label).
• Children and young people. In the UK it is for adults only: "Do not give it to children under 18 years" (NHS). In the US it is licensed as an add-on for seizures from 1 month of age.
• Pregnancy. A Nordic registry study of more than 2,700 first-trimester exposures found major malformations in 5.9% against 4.1% of unexposed babies. After adjustment the prevalence ratio was 1.14 (not significant) against unexposed babies, but 1.29 against lamotrigine and 1.39 against duloxetine, which were significant. The MHRA advises avoiding pregabalin in pregnancy "unless clearly necessary" and using effective contraception (MHRA 2022). If you become pregnant, do not stop suddenly; talk to your doctor. The NHS recommends high-dose folic acid (5 mg) and notes that newborns may need monitoring for withdrawal (NHS).
• Breastfeeding. Pregabalin passes into breast milk. The FDA label estimates the infant dose at about 7% of the maternal dose on a weight basis, and the US label states that breastfeeding "is not recommended" while taking it (FDA label). UK and US advice differ here. The NHS says it can usually be taken while breastfeeding a healthy baby after discussion, and advises watching for unusual sleepiness and not sharing a bed with the baby (NHS).
• People with myoclonic or absence seizures, which it can worsen (NICE NG217).

Dosage and how to take it

How to take it. Take it with or without food, but the same way each day. Swallow tablets or capsules whole. If you miss a dose, take it when you remember unless your next dose is within 2 hours. Never take a double dose (NHS).

How long before it works. The NHS says "it takes at least a few weeks for pregabalin to work" (NHS). For post-herpetic neuralgia and spinal cord injury pain, the FDA label allows a dose increase if relief is not enough after 2–4 weeks and 2–3 weeks respectively.

How long people stay on it. For nerve pain or anxiety, the NHS says people usually continue for several months after symptoms go, to stop them returning. For epilepsy, treatment may last many years (NHS).

How to stop. Never stop abruptly. The EU and US product information say to taper "over a minimum of 1 week", whatever the reason it was prescribed (EU SmPC, FDA label). One week is a regulatory minimum. After long-term or high-dose use, withdrawal may be dose-related, and your prescriber will plan the reduction with you. NICE refers prescribers to its guideline on medicines associated with dependence or withdrawal symptoms (NICE NG193).

If too much is taken. Go to A&E (or call your local emergency number) if someone has taken more than prescribed and is sleepy, confused, agitated, having a seizure or has passed out (NHS).

Follow the money: who makes it and who funded the evidence

The ownership chain

• Invention: Northwestern University (USA), in chemist Richard Silverman's laboratory. The university sold its initial royalty interest for $700 million in 2007 and its remaining royalty interests in 2014. Its patent and licensing income fell from $357 million in 2014 to $32 million in 2015 (The Daily Northwestern, 2015; a student newspaper, used here only for these figures).
• Developer and originator marketer: Pfizer Inc (New York, USA), through Warner-Lambert and its Parke-Davis division (UK Supreme Court, DailyMed).
• Peak sales: Lyrica earned Pfizer $4,970 million worldwide in 2018 ($3,594 million of it in the US) and $5,065 million in 2017 (Pfizer 2018 Financial Report). After US generics arrived in July 2019, revenue fell to $3,321 million in 2019 (Pfizer 2019 Financial Report).
• Current owner: In 2019 Pfizer placed Lyrica in its Upjohn off-patent business. On 16 November 2020 Upjohn combined with Mylan to form Viatris Inc, headquartered in Canonsburg, Pennsylvania, USA. Viatris reported Lyrica net sales of $487.0 million in 2025, down from $556.5 million in 2023 (Viatris 10-K FY2025). The EU licence is held by Upjohn EESV, Netherlands (EU product information).
• Generics: many manufacturers worldwide now make pregabalin. Generic prices are low, so today's prescribing is not driven by brand revenue the way it was before 2019.

Who funded the evidence

• Efficacy trials: the Cochrane epilepsy review found every included trial was Pfizer-sponsored (Panebianco 2022). The Cochrane neuropathic-pain review's study tables record Pfizer sponsorship or funding for most trials, and the review notes large amounts of unpublished data (Derry 2019).
• Independent tests: the Australian government-funded PRECISE trial (sciatica) was negative (NEJM 2017). The Wellcome/Swedish-funded registry study (BMJ 2019) found harms signals.
• Safety study funded by the manufacturer that found harm: the Nordic pregnancy study that led to the 2022 MHRA warning was funded by Pfizer Inc, according to its ENCePP register entry. It was a regulator-requested post-authorisation study, and it found a small malformation signal. This shows that manufacturer funding does not always produce favourable results.
• Regulators: both the MHRA and FDA rely heavily on industry fees. The FDA's human drug review was about 77% fee-funded in FY2025, and the EMA's budget is about 91.5% fee-funded (FDA PDUFA report, EMA funding). Even so, the EMA refused the fibromyalgia licence, and all three regulators added respiratory warnings.

Documented integrity and legal events

• 2009 US settlement ($2.3 billion total): Pfizer agreed to pay $1 billion "to resolve allegations under the civil False Claims Act" that it illegally promoted four drugs, including Lyrica, for uses that were not medically accepted indications. The settlement also resolved allegations of kickbacks to health care providers. The criminal guilty plea in the same settlement concerned a different drug, Bextra. Pfizer also entered a corporate integrity agreement (US Department of Justice, 2 September 2009). These were allegations resolved by settlement, not findings at trial.
• 2018 UK patent ruling: Warner-Lambert (Pfizer) held a second patent covering pregabalin for pain. The UK Supreme Court held that the claims to pain and neuropathic pain "fail for insufficiency": the patent's disclosure supported inflammatory pain but not neuropathic pain (UK Supreme Court press summary, [2018] UKSC 56). This was a patent-law ruling about what the patent document disclosed in 1996, not a ruling on whether pregabalin works for nerve pain today.
• Regulatory divergence: the EMA refused fibromyalgia in 2009, while the US licence includes it. The US licence does not include GAD, while the UK/EU licence does.

None of this means pregabalin does not work. The independent Cochrane reviews confirm benefits in specific conditions. What it does mean is that the evidence base was largely built by the company that sold the drug, that unpublished data remain, and that its US marketing was the subject of a large federal settlement. Readers should weigh the independent trial results (PRECISE) and the independent harm studies (Molero, regulators) accordingly.

Related research

• Stress, anxiety and depression: evidence-based prevention guide
• Supplements for stress, anxiety and depression: the evidence
• Sertraline: independent evidence, NICE's first-choice drug for GAD
• Escitalopram · Duloxetine · Venlafaxine
• Amitriptyline, another NICE first-choice option for nerve pain
• Diazepam and benzodiazepines: why they are not for long-term anxiety
• Sleep prevention guide · Sleep supplements evidence
• Magnesium · L-theanine · Ashwagandha · St John's wort

Frequently asked questions

Sources and funding notes

• NHS: Pregabalin (medicine pages, reviewed 20 January 2026): UK government health service; no manufacturer funding.
• NICE CG113: Generalised anxiety disorder and panic disorder in adults: UK public body (DHSC grant-in-aid plus some company appraisal fees).
• NICE CG173: Neuropathic pain in adults: as above.
• NICE NG193: Chronic pain: as above.
• NICE NG217: Epilepsies, chapter 5: as above.
• MHRA Drug Safety Update, April 2019: rescheduling: UK regulator (partly fee-funded).
• MHRA Drug Safety Update, February 2021: severe respiratory depression: UK regulator.
• MHRA Drug Safety Update, April 2022: pregnancy: UK regulator. The underlying Nordic study was Pfizer-funded (ENCePP entry).
• PHE / NHS England: advice on misuse risk (2014): UK public bodies.
• FDA Drug Safety Communication, 19 December 2019 (archived copy): US regulator; drug review about 77% industry-fee funded.
• Lyrica US prescribing information (Viatris, revised 04/2025), via DailyMed: manufacturer-written, FDA-approved.
• EMA: Lyrica EPAR and EU product information: EU regulator (about 91.5% fee-funded); efficacy data from sponsor trials.
• EMA: refusal of fibromyalgia indication (23 July 2009): EU regulator.
• Derry et al., Pregabalin for neuropathic pain in adults, Cochrane 2019 (full text): UK academic; NIHR Cochrane infrastructure and Oxford Pain Research funds; one author declared honoraria from Mundipharma and Grünenthal, another from RB, Novartis and others; included trials mostly Pfizer-sponsored.
• Derry et al., Pregabalin for pain in fibromyalgia, Cochrane 2016: UK academic; one author declared grants or honoraria from RB, Grünenthal, Menarini and Novartis.
• Panebianco et al., Pregabalin add-on for drug-resistant focal epilepsy, Cochrane 2022: UK academic (Liverpool); NIHR-funded; all included trials Pfizer-sponsored.
• Mathieson et al., Trial of Pregabalin for Acute and Chronic Sciatica, NEJM 2017: Australian academic; funded by NHMRC.
• Slee et al., Pharmacological treatments for GAD, Lancet 2019: UK academic (UCL); "No funding was received"; author declarations not verified by us.
• Generoso et al., Pregabalin for GAD meta-analysis, Int Clin Psychopharmacol 2017: Brazil/Austria/USA academic; funding and conflicts not shown in the abstract record.
• Molero et al., Gabapentinoids and adverse outcomes, BMJ 2019 (full text): Sweden/UK academic; Wellcome Trust and Swedish research councils; one author declared unrelated Shire and Evolan ties.
• Evoy et al., Abuse and misuse of pregabalin and gabapentin, Drugs 2017: US academic (University of Texas); funding not shown in the abstract record.
• Gomes et al., Pregabalin and the risk for opioid-related death, Ann Intern Med 2018: Canadian academic (ICES); research letter with no abstract; its finding is cited here via the MHRA.
• Gomes et al., Gabapentin, opioids and opioid-related death, PLoS Med 2017: Canadian academic; some authors declared honoraria from several drug makers, including Pfizer, outside the submitted work.
• AGS Beers Criteria 2023, J Am Geriatr Soc: US professional society; "no sponsor"; minor declared consulting ties, none to pregabalin makers.
• ONS: Deaths related to drug poisoning in England and Wales, 2024 registrations: UK national statistics body.
• US Department of Justice: Pfizer $2.3 billion settlement, 2 September 2009 (archived): US government.
• UK Supreme Court: Warner-Lambert v Generics (UK) Ltd [2018] UKSC 56, press summary: UK court.
• Pfizer 2018 Financial Report (SEC) and Pfizer 2019 Financial Report (SEC): company filings; figures are audited, but framing is the company's own.
• Viatris 10-K, FY2025 (SEC): company filing.
• FDA PDUFA financial report FY2025 and EMA funding page: regulator funding disclosures.
• The Daily Northwestern (2015): Lyrica royalties: student newspaper; lower-tier source, used only for royalty figures.

Last reviewed: 26 September 2026. Educational information from Pure City Research, not medical advice. Medicine decisions belong with a qualified prescriber.